Oligonucleotide-modified Nuclotides

(1)

Oligonucleotide-modified Nuclotides

Dissertation

Zur Erlangung des akademischen Grades des Doktors der Naturwissenschaften

(Dr. rer. nat.)

an der Universität Konstanz

Mathematisch-Naturwissenschaftliche Sektion Fachbereich Chemie

vorgelegt von Anna-Lena Steck

Tag der mündlichen Prüfung: 13.12.2013

1. Referent: Herr Prof. Dr. Andreas Marx 2. Referent: Herr Prof. Dr. Jörg Hartig

Konstanzer Online-Publikations-System (KOPS) URL: http://nbn-resolving.de/urn:nbn:de:bsz:352-288756

(2)

II

(3)

III

(4)

IV

(5)

V Publications

Parts of this thesis are published in:

Barcoded nucleotides. A. Baccaro,⁺ A.-L. Steck,⁺ and A. Marx, Angew. Chem., Int. Ed. Engl., 2012, 51(1): 254-257.

Structures of KlenTaq DNA Polymerase Caught While Incorporating C5-Modified

Pyrimidine and C7-Modified 7-Deazapurine Nucleoside Triphosphates. K. Bergen,⁺ A.-L.

Steck,⁺S. Strütt, A. Baccaro, W. Welte, K. Diederichs, and A. Marx J. Am. Chem. Soc. 2012, 134, 11840-11843. (Cover article)

+ These authors contributed equally to this work.

(6)

VI

(7)

VII

VIII Table of contents

3.2.3 Conclusion...63

4 Summary ... 65

5 Zusammenfassung... 69

6 Materials and Methods ... 73

6.1 General ... 73

6.1.1 Chemicals and solvents...73

6.1.2 Electrospray ionization mass spectrometry (ESI-MS)...73

6.1.3 Nuclear magnetic resonance (NMR) ...73

6.1.4 Chromatography...74

6.1.4.1 Thin layer chromatography (TLC) ...74

6.1.4.2 Flash chromatography ...74

6.1.4.3 Middle pressure liquid chromatography (MPLC) ...74

6.1.4.4 High pressure liquid chromatography (HPLC)...74

6.1.4.5 Fast Performance Liquid Chromatography (FPLC)...74

6.2 Synthesis of modified nucleotides ... 75

6.2.1 Synthesis of 2’-deoxyadenosine analogs...75

6.2.2 Synthesis of 2’-deoxyguanosine analogs...77

6.3 Biochemical Experiments... 78

6.3.1 Buffers and solutions ...78

6.3.2 Enzymes...78

6.3.3 Nucleotides...79

6.3.4 Oligonucleotides and modified oligonucleotides ...79

6.3.5 5’-Radioactive labeling of ODNs using [γ-³²P]ATP ...79

6.3.6 Radioactive denaturating gel electrophoresis (PAGE) ...79

6.3.7 Non-radioactive denaturating gel electrophoresis...80

6.3.8 Preparative gel electrophoresis ...80

6.3.9 Ethanol precipitation...80

6.3.10 Circular dichroism (CD) measurements...80

6.3.11 Circularization of the rolling circle template ...81

6.3.12 Synthesis of activated oligodeoxynucleotides (ODN) ...81

6.3.13 Synthesis of ODN-modified nucleotides...81

6.3.14 Quantification of ODNs and calculation of extinction coefficients...81

6.3.15 DNA sequences ...82

6.3.16 Primer extension reaction...83

6.3.17 Competition experiments using amine-modified nucleotides ...83

6.3.18 Competition experiments using ODN- and amine-modified nucleotides...83

6.4 DNA microarray experiments ... 84

6.4.1 Nanospotter ...84

6.4.2 Spotting and immobilization of DNA oligomers (primer strands)...84

6.5 Reactions on streptavidin sepharose ... 86

(9)

Table of contents IX

7 Abbreviations...89

8 References ...93

Eidesstattliche Erklärung... 105

Danksagung ... 107

(10)

X

(11)

1 1 Introduction

60 years past, since R. Franklin, M. Wilkins, J. Watson and F. Crick discovered the structure of DNA. Their publications in 1953 had a major impact in the field of biology and offered great opportunities.^[1-5] Particularly in the field of molecular biology, development began with an astonishing pace, culminating in the sequencing of the human genome in 2001.^{[6, 7]}

As 99.9% of the human genome is shared within all individuals, it is often not necessary to sequence the complete human genome to acquire important information.^[6] The detection of particular sequence differences between individuals is often sufficient. Most abundant genetic variations are single nucleotide polymorphisms (SNPs), in many cases they are responsible for a predisposition to certain diseases and different drug efficiencies in individuals.^[8-10] Especially variations within the coding region of an enzyme are known to influence the phenotype. These variations can change the amino acid sequence of the expressed protein leading to modified structures, activities and functions of the respective proteins. For instance a single nucleotide variation within the sequence coding for the haemoglobin protein leads to an amino acid change in the hemoglobin protein giving the red blood cells a sickle shape form under low-oxygen conditions (sickle-cell disease).^[11]

Nowadays, it is even possible to determine someone’s ancestry by the help of the SNP pattern.^{[12, 13]} Consequently, the knowledge of single nucleotide polymorphisms opens the way for personalized medicine and anticipatory disease treatment.

Several techniques have been invented so far to detect SNPs. All known techniques have in common that they need expensive lab equipment and highly trained stuff. Although, the sequencing prices decreased dramatically during the last years, making the detection of SNPs not only possible but also affordable, but an easy point-of-care for the detection of SNPs is still missing.

1.1 DNA

DNA is the information storage system of nearly all living organisms (except of RNA viruses). The biopolymer consists of four different 2’-deoxyribonucleotide units:

2’-deoxyguanosine monophosphate (dGMP), 2’-deoxyadenosine monophosphate (dAMP),

(12)

2 Introduction

2’-deoxythymidine monophosphate (dTMP) and 2’-deoxycytidine monophosphate (dCMP) (Figure 1). A nucleotide is composted of a nitrogenous base connected to a 5’-phosphate- 2’-deoxyribose and is linked to the next nucleotide via a phosphodiester linkage.

Furthermore, the nucleobases can be classified into pyrimidine derivatives (thymine and cytosine) and purine derivatives (adenine and guanine) (Figure 1B). The lack of the 2’-hydroxylgroup of the ribose-moiety makes DNA more resistant to hydrolysis compared to RNA, making it to the superior molecule for long-term storage of the genetic information.

According to the Watson-Crick rules, the nitrogenous base adenine forms a base pair with thymine and guanine pairs with cytosine (Figure 1C). So, Watson-Crick base pairing is specific, predictable and enables the self-assembly of oligodeoxynucleotides. By canonical Watson-Crick base pairing two antiparallel strands form a characteristic geometrically well-defined double helical structure with major and minor groove (Figure 1D). The double helical structure is stabilized by the Watson-Crick hydrogen bonding between the nucleobases of the two strands, π-stacking- and hydrophobic-interactions of the aromatic nucleobases.^[14] Whereas double stranded DNA forms a helical structure, single-stranded DNA can form different secondary structures (e.g. G-quadruplexes, see chapter 1.1.1).

Figure 1 A) Structure of the basic nucleotide building block. B) Structures of the four natural DNA nucleobases: guanine (G), adenine (A), thymine (T) and cytosine (C). C) Watson-Crick base pairing of A with T and G with C. Hydrogen bonds are illustrated in light-blue. D) Double-helical structure of DNA (B-type structure) (PDB-code 1BNA).¹

The complete human genome comprises 3.2 billion base pairs. The sequence of these nucleotides encodes the information for around 23 thousand genes. Every gene codes for a protein that is expressed via transcription and translation. During translation, sequences of three nucleotides in a row (triplet codons) specify the amino acid sequence of the

1 All images of pdb-files were prepared using PYMOL (DeLano Scientific).

(13)

Introduction 3

respective protein.^[15] Every protein has a special function within the living cell and mutations within the DNA sequence can lead to inoperative or dysfunctional proteins. This can lead to serious diseases (see also chapter 1.1.2). On the other hand, this inaccuracy also facilitates evolution.

Before cell-division, the cell must copy its genetic information. The process of DNA duplication is termed replication and is catalyzed by DNA-dependent DNA polymerases. It is also possible to perform DNA replication in vitro by incubation of a primer/template complex with a DNA polymerase and nucleotides. Since the development of the polymerase chain reaction (PCR) by K. Mullis, it is also possible to exponentially amplify a DNA fragment in vitro.^[16]

DNA is also an interesting tool for biotechnological applications, because of its outstanding properties like self-assembly and the formation of a defined helical structure according to the hybridization specificity. These properties have already been used to create several geometrical structures with DNA, like cubes^[17]. In addition, the simple exponential amplification of DNA in vitro by PCR enhances the applicability of DNA for biotechnical applications.

1.1.1 G-quadruplexes

Single-stranded G-rich DNA strands are capable to form G-quadruplex structures.^[18]

Thereby four guanine bases form a planar structure called G-tetrad (Figure 2). The guanine bases within one G-tetrad associate through Hoogsteen hydrogen bonding^[19]. In contrast to Watson-Crick base pairing, O6 and N7 of the guanine base are proton acceptors whereas N1 and the C2 amino group are proton donors. The G-tetrads can stack on each other by π-π stacking and form a G-quadruplex structure (Figure 2).

Figure 2 Structure of a G-quadruplex. Left: Illustration of a G-tetrad, which is build by four guanine bases. The guanine bases interact through Hoogsteen hydrogen bonding (shown in light-blue).

Right: G-tetrads can stack on each other by π-π stacking and form a G-quadruplex structure.

(14)

4 Introduction

The G-quadruplex structure is further stabilized by a cation.^{[20, 21]} The exact position of this cation can vary dependent on the structure and the cation. Especially potassium ions, which allow the best stabilization of the G-quadruplex structure, are located between two G-tetrads, interacting with the eight oxygen ions of the two G-tetrads (Figure 3).

Figure 3 Crystal structure of a G-quadruplex (PDB-code 1KF1) with potassium ions. Potassium ions are shown in green, carbon in blue, oxygen in red, nitrogen in dark-blue. A) Top view of a G-tetrad with potassium ions. B) Side view of the human telomeric G-quadruplex with potassium ions. The potassium ions are located between the G-tetrads. Loop molecules are not depicted. C) Top view of the human telomeric G-quadruplex.

G-quadruplexes can be formed by DNA, RNA, PNA and LNA sequences.^{[22, 23]} The best studied G-quadruplex sequence d(GGTTAG)x is the human telomeric repeat sequence.

G-quadruplexes are also known to catalyze enantioselective Friedel-Crafts reactions and Diels-Alder reactions with modest enantioselectivities.^[24-26]. Additionally, one of the most popular DNAzyme is based on a G-quadruplex – the hemin/G-quadruplex DNAzyme (see also chapter 1.1.1.1).

1.1.1.1 The hemin/G-quadruplex DNAzyme

DNA G-quadruplexe sequences are also known to have peroxidase-mimicking catalytic activity and are classified as DNAzymes. DNAzymes are catalytic active oligodeoxynucleotides discovered in 1994 by R. Breaker and G. Joyce.^[27] They found that a short oligodeoxynucleotide sequence catalyses the lead ion dependent cleavage of a RNA substrate. In contrast to Ribozymes^{[28, 29]} (RNAzymes, catalytic active RNA), which occur in nature, DNAzymes have not been found in nature so far.^[30] Typically, they are discovered by in vitro selection (e.g. by the help of a systematic enrichment of ligands by exponential amplification (SELEX) process)[27, 31, 32]. Compared to Ribozymes, DNAzymes are less expensive to produce and more stable against hydrolysis, making them an expedient tool for biotechnical applications.^[33] DNA bears also the advantage of a direct amplification by PCR without the need of a reverse transcription step as it is necessary for RNA. By now several DNAzymes that catalyze a huge variety of reactions are known.[27, 31, 34-37]

(15)

Introduction 5

One popular DNAzyme is the hemin/G-quadruplex DNAzyme (Figure 4). This DNAzyme consists of hemin and a guanine-rich single-stranded DNA, which forms a G-quadruplex structure.^[38-46] This complex can effectively catalyze the H2O2-mediated oxidation of ABTS^2- to generate the colored radical product ABTS^•^-. The formation of the ABTS^•^- radical can be detected either by absorbance measurement or by naked eye.

Figure 4 Schematic depiction of the hemin/G-quadruplex DNAzyme system. The complexation of hemin (purple) with a guanine-rich DNA strand (blue) yields a G-quadruplex structure that exhibits peroxidase activity. This complex can effectively catalyze the H2O2-mediated oxidation of ABTS^2- to generate the colored radical product ABTS^•-. The formation of the ABTS^•- radical can be detected either by naked eye or by absorbance measurement.

The possibility of naked eye detection makes this DNAzyme to a versatile tool in biotechnical applications. And indeed over the past years, the peroxidase-mimicking hemin/G-quadruplex DNAzyme has been used as chameleonic label for various colorimetric or chemiluminescent assays.^[38-44] For instance, I. Willner and coworker developed an analytical platform for monitoring telomerase activity in cells.^[39] In 2010, I.

Willner and coworker found that the hemin/G-quadruplex DNAzyme also acts as NADH oxidase and NADH peroxidase mimicking enzyme.^{[33, 47]} Recently, H. Abe et al. synthesized a variation of this DNAzyme which is soluble in water and in most organic solvents by coupling a polyethylene glycol unit to the 5’-end of a G-quadruplex forming oligonucleotide.^[48] The hemin/G-quadruplex structure is also able to promote a chemiluminescence resonance energy-transfer process or quenching of luminescence of quantum dots^[49-52] and the structure bioelectrocatalyses the electrocatalysed reduction of H2O2[53, 54].

(16)

6 Introduction

1.1.2 Single-nucleotide polymorphisms (SNP)

The human genome comprises 3.2 billion DNA base pairs, whereas the nucleotide sequences of individuals vary by less than 0.1%.^[6] The most frequent genome variations are single-nucleotide polymorphisms (SNPs).^[55] SNPs are single nucleotide changes at a specific position in the genome.^[56] Together with copy number variations, SNPs are mostly responsible for the uniqueness of every human being.^{[57, 58]} SNPs can occur all over the genome; within coding regions of genes, non-coding areas or between these regions.

Dependent on the position of the variation in the genome, it can lead to modified structures, activities and functions of proteins. It is known that DNA variations contribute to diseases (such as cancer, sickle cell anaemia^[11] or Alzheimer’s disease ^[59-61]) and predisposition to side effect of drugs are linked to SNPs.^{[8, 9]} It is already possible to adapt a drug therapy to the genetic basis of a patient to avoid side effects. Before an intake of some drugs a genetic test is recommended or even compulsory. One example is the drug Abacavir (Ziagen^®, GlaxoSmithKline), a nucleoside analog reverse transcriptase inhibitor to treat HIV and AIDS patients. One of the main side effects of Abacavir is severe hypersensitivity including e.g. fever, fatigue, gastrointestinal symptoms and respiratory symptoms.^[62] The hypersensitivity is strongly connected to the presence of the HLA- B*57:01 allel for which testing is compulsory before an intake of this antiretroviral drug.^[63-68] The side effect is probably explained by the binding of Abacavir to the peptide- binding groove of HLA-B*57:01, altering the spectrum of peptides that bind to this allel.^[69,

70]

Several methods for SNP detection are known. All known techniques have in common that they need expensive lab equipment and highly trained stuff. They are based on dye- labeled nucleotide incorporation or sequence-specific hybridization, for example.

Although, the sequencing prices decreased dramatically during the last years; so sequencing becomes more and more affordable for SNP detection.

1.2 DNA polymerases

DNA polymerases are the key enzymes in DNA synthesis. These enzymes do not only play a central role in DNA replication by catalyzing the elongation of the primer strand by nucleotides in a template dependent manner, they also play a significant role in DNA damage repair and recombination.^[71] The intrinsic selectivity of DNA polymerases is of central importance for the accuracy of DNA replication.^[72] The accuracy of DNA polymerases results not only from correct hydrogen bond formation, other effects turned out to be much more relevant. Insight into the selection of the right nucleotide was provided by E. Kool and coworkers.^[73-76] They synthesized an artificial base pair, which is not able to form Watson-Crick hydrogen bonds, and explored the incorporation properties. It turned out that geometric complementarity in the active site of the DNA

(17)

Introduction 7

polymerase of the nascent nucleobase pair is fundamental. Nascent nucleotide base pairs that fit well in the active centre are well accepted whereas incorrect 2’-desoxynucleoside triphosphates are processed with lower efficiency.^[73] Known features that contribute to the selectivity are the geometry of the active centre, the steric demand of the incoming nucleoside triphosphate, π-π interaction, solvent effects and interaction of the polymerase and the minor groove of B-DNA.[74, 75, 77-79]

The structure of most DNA polymerases can be compared to a right hand. It comprises the domains: fingers, thumb and palm (Figure 5). In the palm domain the active center is located; this domain comprises amino-acid residues that allow the coordination of the bivalent ions.^{[80, 81]} The fingers domain binds the template strand across from the primer.^[81] The thumb domain interacts with the minor groove of the DNA synthesis product.^[81]

By evolution every DNA polymerase has been tailored to its specific function in the cell.

Consequently there are replicative polymerases (e.g. DNA polymerases δ and ε) with a high selectivity and fidelity and polymerases (e.g. DNA polymerase β) involved in DNA repair of damaged or mismatched bases with low fidelity on undamaged templates.

Figure 5 Crystal structures of KlenTaq DNA polymerase in the open (left, PDB-code 4KTQ) and closed (right, PDB-code 3KTQ) conformations. The O-helix undergoes a rotation from the open to the closed conformation. The structures of the enzyme are comparable to a right hand consisting of the finger domain (orange), the palm domain (blue) and the thumb domain (green). The surface of the DNA polymerase is shown semitransparent (grey). The primer and the template are shown in magenta and light-magenta. The incoming nucleotide is shown in black.

On the basis of structural similarities and sequence homology DNA polymerases are classified into seven different DNA polymerase families (A, B, C, D, X, Y, RT).^[82-85] For example, DNA polymerases belonging to family A are mostly homolog to E. coli DNA polymerase I.^[71] Prominent members of the A family are Taq DNA polymerase, Klenow

(18)

8 Introduction

fragment (KF) DNA polymerase (see chapter 1.2.3) and T7 DNA polymerase. Vent and Pfu DNA polymerase are members of the B family and are mostly homolog to E. coli DNA polymerase II.^[71] Members of family Y (e.g. DNA polymerase IV (DPO4)) are known as translesion synthesis DNA polymerases.^[71] They have a low fidelity when replicating undamaged templates but they are able to bypass DNA lesions.^[71]

Thermostable DNA polymerases spearheaded by Taq DNA polymerase revolutionized the biotechnology field by easing the practicability of PCR. Nowadays, DNA polymerases became an indispensable tool in biotechnological applications, due to their widespread applicability in PCR and DNA sequencing technologies.

In chapter 1.2.1.1 the synthesis cycle of DNA polymerases is exemplified by KlenTaq DNA polymerase.

1.2.1 KlenTaq DNA polymerase

The A-family member KlenTaq DNA polymerase is a truncated version of the thermostable Taq DNA polymerase. Taq DNA polymerase was isolated from the thermophilic bacterium Thermus acquaticus, which was first discovered in a geyser of the Yellowstone National Park by T. Brock.^[86] Due to the thermal stability of Taq DNA polymerase (activity half-life of approximately 45-50 min at 95°C and 9 min at 97.5°C)^[87], its processive speed and fidelity, it is commonly used in PCR.^[16]

Apart from the lack of a 3’-5’ exonuclease proofreading activity, the sequence of Taq DNA polymerase is very homologue to E. coli DNA polymerase I. In analogy to the Klenow fragment of the E. coli DNA polymerase I, Taq DNA polymerase without the 5’-3’

exonuclease function is called KlenTaq DNA polymerase or large fragment of Taq DNA polymerase I.^[88] Thus, KlenTaq DNA polymerase conveys only DNA synthesis activity. The certainty that KlenTaq DNA polymerase lacks both exonuclease activities makes it a prominent DNA polymerase for the incorporation of modified nucleotides. For instance, Wang et al. showed that KlenTaq DNA polymerase is able to incorporate diamondoid-modified nucleotides.^[89]

Since thermostable DNA polymerases are used in many diagnostic and biotechnological applications, there is great interest in variants with customized properties. Hence, several variants of KlenTaq DNA polymerase have been evolved including mutants with substantial reverse-transcriptase activity^[90] or with an ability to amplify from highly UV-damaged DNA^[91].

(19)

Introduction 9

1.2.1.1 Structure and function of KlenTaq DNA polymerase

In 1998 G. Waksman and coworkers succeeded in crystallizing and solving the open and the closed ternary complexes of KlenTaq DNA polymerase.^[92] The three-dimensional structures show a typical structure of a DNA polymerase comparable to a right hand. Thus, it comprises the basic domains: fingers, thumb and palm (Figure 5). In accordance to the general reaction mechanism of DNA polymerases, KlenTaq DNA polymerase adopts an open conformation (binary status, semi-closed hand) binding the primer/template complex (Figure 6). Upon correct nucleotide binding, especially the finger domain undergoes a vast conformational change, switching from the open to the catalytically active closed conformation (ternary status closed, closed hand) (Figure 6B). During this conformational change the active site is formed and the nucleotide is bound tightly. The 5’

α-phosphate group of the nucleoside triphosphate is now in an ideal position to be nucleophilic attacked by the 3’ hydroxyl group of the primer. After the nucleotide is covalently bound to the primer and the pyrophosphate is released the open conformation is adapted to allow translocation for further nucleotide binding or dissociation from the DNA.

Figure 6 DNA polymerase catalytic cycle of nucleotide insertion. A) Simplified depiction of selected steps of the DNA polymerase catalytic cycle. The DNA primer/template complex is bound to the polymerase in the open conformation. The closed polymerase conformation is stabilized upon nucleotide binding. After nucleotide insertion, pyrophosphate (PP) is released triggering the opening of the polymerase and translocation of the DNA strand. The conformational state of the polymerase is shown in green. Figure modified from J. Pata et al.^[80] B) Open (left, PDB-code 4KTQ) and closed (right, PDB-code 3KTQ) complex of KlenTaq DNA polymerase. The primer strand is shown in light-blue the template strand in blue, the templating base in white, the incoming

nucleotide in black and the magnesium ions magenta. The O-helix is highlighted in green. The black arrow shows the rotation of the O-helix. The surface of the polymerase is shown semitransparent (grey).

(20)

10 Introduction

During this reaction process two divalent cations (usually magnesium ions) are involved.^[93-96] They coordinate with two aspartate side-chains in the thumb-domain, with the phosphate groups of the incoming nucleoside triphosphate and with the 3’-OH group of the primer end (Figure 7). Thereby one metal ion coordinates the 3’-OH group of the primer end and probably decreases the pKa value, making the attack to the α-phosphate feasible. A second divalent metal ion orients the triphosphate residue in the active site.^[97]

Figure 7 A) Schematic depiction of the extended two-metal-ion mechanism of nucleotidyl transfer.

Depicted are the active center of a polymerase, two divalent ions (magnesium ions, magenta), water molecules (black dots), the primer end (light-blue) and the incoming nucleoside triphosphate (black). One divalent metal ion is coordinated by three oxygen molecules of the triphosphate group of the incoming nucleoside triphosphate, water molecules (supposed) and an aspartate residue located in motif A of the polymerase (green). The second divalent metal ion is coordinated by the 3’-hydroxyl group of the primer end (light-blue), the α-phosphate group of the nucleoside triphosphate and widely conserved aspartate residues located in the structural motifs A and C (green) and water molecules (supposed). An acidic amino acid (A) is proposed to protonate the pyrophosphate leaving group. A basic amino acid (B) is proposed to provoke the deprotonation of the 3’-hydroxyl group. The dark-blue arrow indicates the nucleophilic attack of the 3’-hydroxyl group and the dashed dark-blue lines indicate the transition state. The figure was prepared according to C. Castro et al.^[97] B) Interactions of the two Mg²⁺-ions (purple) in the active centre of KlenTaq DNA polymerase (PDB-code 3KTQ). The primer end is depicted in light-blue and the incoming nucleoside triphosphate in black.

1.2.2 Therminator DNA polymerase

Therminator DNA polymerase is a variant of the thermostable 9°N DNA polymerase with an enhanced ability to incorporate modified nucleotides.^[98] 9°N DNA polymerase was found in the hyperthermophilic maritime archaeon Thermococcus species 9°N-7.^[99] The difference between Therminator and 9°N DNA polymerase are three mutations (D141A / E143A / A485L).[100, 101] At position 485 the amino acid alanin is changed to leucin. Both amino acids have an unpolar side chain, but leucin is sterically more demanding. Although the mutation is located within the α-helix and not within the active centre, it affects a change in the substrate specificity. The two additional mutations inactivate the 3’-5’

(21)

Introduction 11

exonuclease activity of the DNA polymerase, which is useful for the synthesis of modified DNA.

With its enhanced substrate spectrum and its thermostability, Therminator DNA polymerase is often the DNA polymerase of choice for the incorporation of modified nucleotides. For example, J. Szostak and coworkers found that Therminator DNA polymerase is able to catalyze the synthesis of (3’–2’) α-L-threose nucleic acid (TNA) by substituting all four natural nucleotides with tNTPs.^[102] Studies of S. Obeid et al. showed that Therminator DNA polymerase is able to adjacent incorporate eleven spinlabeled nucleotides, resulting in one complete DNA helix turn equipped with spin-labels (see also section 1.3.2).^[103]

1.2.3 KF (exo^-) DNA polymerase

Is the E. coli DNA polymerase I enzymatically cleaved by the protease subtilisin two fragments are generated. The obtained smaller N-terminal fragment consists of the 5’-3’

nuclease activity and the larger C-terminal unit comprise the polymerase activity and the 3’-5’ exonucleoase proofreading function. The larger fragment was first described by H.

Klenow and is named Klenow fragment (KF).^[104] The Klenow Fragment was originally used in PCR reactions but due to the thermolability of the enzyme, it had to be added after each heat denaturating step at 95°C.^[16] So, it was replaced by thermostable DNA polymerases such as Taq DNA polymerase.^[16]

For the incorporation of modified nucleotides DNA polymerases lacking the 3’-5’

exonucleoase proofreading function are preferred to prevent removal of the incorporated modified nucleotide.^[101] To generate the KF (exo^-) DNA polymerase, two mutations (D355A / E357A) were introduced which abolish the 3´-5´ exonuclease activity.^[105]

The KF (exo^-) DNA polymerase had been successfully used for rolling circle amplification reactions.^{[106, 107]} Thereby a primer DNA strand is extended by multiple copies of the same sequence using a circular template (Figure 8).[108, 109] In contrast to PCR a heat denaturation step is not needed and the reaction can be done under isothermal conditions.

(22)

12 Introduction

Figure 8 Rolling circle amplification. A primer DNA strand is hybridized to the circular template (dark-blue). The DNA polymerase starts with the template-directed elongation of the primer strand. After one round of elongation the primer strand is displaced and the extension of the RCA product (light-blue) continuous. The RCA product consists of multiple copies of the same sequence.

1.3 Synthesis of modified DNA

Natural DNA consists of four building blocks; the nucleotides dGMP, dAMP, dCMP and dTMP connected via phosphodiester linkage. To change the properties of natural DNA, nucleotides with miscellaneous properties are needed for biotechnological applications like sequencing^[110-113]. To create modified DNA basically three different approaches are feasible: the automated DNA synthesis by DNA synthesizers, the enzymatic incorporation by DNA polymerases or post-synthetic labeling.^[114]

Automated DNA synthesis by DNA synthesizers

The automated DNA synthesis on solid support offers a great opportunity to synthesize short DNA strands (up to approximately 100-nt) in moderate time and yield. The modified building blocks can be incorporated into the DNA strand within the synthesis cycle, providing that the modified building block is available as 3’-phosphoramidite derivative (Figure 9A) and is compatible with the chemicals used during the synthesis cycles. Does the modification fulfill these requirements the type and the position of the modification can be feely selected.^[115] Not only nucleoside analogs can be incorporated using the automated solid support synthesis, also activating groups (e.g. C10-carboxy-modifier) or alkyl spacer (e.g. C-18-spacer) can be incorporated. Though, the length of the modified oligonucleotide is restricted due to the fact that the coupling efficiencies of modified phosphoramidite derivates is diminished compared to unmodified building blocks. This has also an effect on the feasible modification density, which is often quite low.

(23)

Introduction 13

Enzymatic DNA synthesis by DNA polymerases

The enzymatic incorporation by DNA polymerases allows the synthesis even of long modified DNA strands for example by using polymerase chain reactions or primer extension reactions (Figure 9B). The modified nucleotide is thereby applied as 5’-triphosphate derivative (Figure 9C). Examples for modified nucleoside triphosphates are depicted in Figure 10 and Figure 12. Although the incorporation of modified nucleotides by DNA polymerases is widely employed, the acceptance of modified nucleotides by a DNA polymerase is often not predictable, making the choice of a suitable DNA polymerase challenging (see also chapter 1.3.1).^[114] The incorporation of non- nucleotide modifications like alkyl spacer or activating groups is not possible.

Figure 9 A) Phosphoramidite building block for the automated DNA synthesis on solid-support. B) Schematic depiction of a primer extension reaction. The DNA polymerase (grey) catalyses the template-directed elongation of the primer DNA strand (dark-blue) by nucleotides (dark-green). C) Nucleoside triphosphate building block for the enzymatic incorporation by DNA polymerases. D) Exemplary depiction of a post-synthetic labeling reaction using click chemistry.

Postsynthetic-labeling approach

An alternative to these two methods is the postsynthetic-labeling approach. Thereby small reactive group (such as an amine or azide groups) are incorporated into a DNA strand either by automated or enzymatic synthesis. Adjacent the desired modification is coupled postsynthetically for example via amide bond formation, Staudinger ligation[116, 117], Click chemistry^[118-121] (Huisgen-Meldal-Sharpless reaction) or Diels-Alder reaction[122, 123]

(Figure 9D). Several modifiers for the incorporation of reactive groups by automated or

(24)

14 Introduction

enzymatic synthesis are commercially available (e.g. N6-(6-azido)hexyl-dATP, C10-carboxy-modifier, amino-modifier-C8-dA). The most challenging part is the purification of the desired modified DNA strand from unreacted DNA and reaction additives like copper.

1.3.1 Enzymatic incorporation of modified nucleotides

The enzymatic incorporation of nucleotides by DNA polymerases allows the synthesis of long modified DNA strands by using PCR or primer extension reaction (Figure 9B). The incorporation of modified nucleotides during PCR is more challenging, as the recently synthesized modified DNA strand serves as template for the next extension step. Is the modified nucleotide analog used instead of the natural nucleotide counterpart a high modification density is achieved, but this can also impinge the reaction progress.

For the acceptance of a modified nucleotide, the position of the modification and the steric demand of the modification play a decisive role. It is important that the Watson-Crick base pairing is not affected by the modification. Modifications can either be introduced at the deoxyribose- or the nucleobase-moiety. Using enzymatic incorporation by DNA polymerases, the 2’ position of the deoxyribose-moiety is not a suitable position due to the discrimination of dNTPs and NTPs by DNA polymerases. Modifying the 1’, 3’ or 5’ position of the deoxyribose moiety is possible but often connected with a protracted synthesis.

Furthermore, these positions are close to the polymerization site and point towards the minor groove of B-DNA resulting in a negative effect on the incorporation efficiencies.^[114]

Modifications at the nucleobase are favored providing that the modifications do not affect the Watson-Crick base pairing. Various studies showed that the C5 positions of pyrimidines and the C7 positions² of 7-deazapurines (see also chapter 1.3.2.1) are the best suited positions as the modifications accommodate well in the major groove of the B-DNA helix.[103, 124-126] The location of the modifications at the C7 positions of 7-deazapurines is preferred compared to the easier accessible C8 positions as steric demanding modifications showed low incorporation efficiencies.[124, 127-129] Practicable and frequently used ways to chemically introduce modifications at these positions are metal-mediated cross-coupling reactions like Sonogashira or Suzuki coupling.

The acceptance of modified nucleotides can be increased by the use of a linker unit. The linker separates the nucleotide moiety from the modification molecule to lengthen the distance between the active site and the modification.

Many base-modified nucleotides are known and have already been successfully incorporated into DNA by various DNA polymerases. For examples see chapter 1.3.2.

2 Purine numbering is used throughout the whole thesis.

(25)

Introduction 15

1.3.2 Modified Nucleotides for enzyme-mediated incorporation

Incorporation of modified nucleotides into a DNA strand alters the properties of the strand. Not only the melting temperature or the helix conformation can change, but also new properties like EPR- or affinity markers can be introduced [103, 107, 130-132]. Several nucleotide analogs have been synthesized so far; a selection is depicted in Figure 10.

For instance, A. R. Kore connected 5-bromo deoxyuridine (BrdU) via a linker moiety to a nucleoside triphosphate (Figure 10d).^[131] Before BrdU was incorporated into the DNA of proliferating cells to study the cell-cycle status or the viability of cells. Incorporated BrdU was detected with specific antibodies. The drawback of this method was that the binding of the specific antibody to BrdU required the denaturation of the DNA duplex under harsh heat and acidic conditions.^[133] The BrdU derivative synthesized by A. R. Kore enables a detection without the denaturation of the DNA, as BrdU is located in the major groove of B-DNA and thus sterically accessible by the antibody.^[131] This modified nucleotide allows an easier detection of cells that are actively replicating their DNA.

Figure 10 Modified dTTP derivatives for DNA polymerase incorporation. Examples for (a) ferrocene-, (b) spin-label-, (c) dendrimeric-modifications and (d, e) marker molecule- modifications.[103, 107, 130-132]

1.3.2.1 Modified purine derivatives

8-substituted purines nucleotides have been shown to be poor substrates for DNA polymerases, as they destabilize dsDNA secondary structure.[124, 127-129] A well suited position for purine modifications is position 7 of 7-deazapurines (Figure 11A), as modifications at this position accommodate well in the major groove of DNA.^[124] The structure of 7-deazapurines mimics the structure of purines, so that they can replace purines even in enzyme-mediated reactions. 7-deazapurines are the most extensively studied purine analogs and several preparation routes are known.^[134] The perfect

(26)

16 Introduction

precursor molecules for modified 7-deazapurines are 7-halogenated 7-deazapurines (Figure 11B/C). 7-halogenated derivatives can be further used for metal-mediated cross-coupling reactions like Sonogashira or Suzuki coupling.

Figure 11 A) Depiction of purine (left) and 7-deazapurine (right). Purine consists of an imidazole ring fused to a pyrimidine ring. The purine numbering is depicted in blue. B) Structure of 7-deaza- 2’-deoxy-7-iodoadenosine. C) Depiction of 7-deaza-2’-deoxy-7-guanosine.

In the following selected examples for modified 7-deazaadenosine triphosphates are given. Recently, M. Hocek and coworkers published the synthesis and incorporation of a terpyridine-modified dATP analog.^[135] The terpyridine moiety is thereby attached to the 7-deazaposition via a long and flexible octadiyne linker (Figure 12a).^[135] After the enzyme-mediated incorporation by Pwo DNA polymerase the obtained modified reaction products can post-synthetically complex divalent metal ions.^[135] To perform postsynthetic Staudinger ligation, Weisbrod et al. synthesized a dATP analog that directly functionalize DNA with azide groups that can react with phosphines (Figure 12e).^[136]

Several other C7-modified dATPs have been synthesized to modify DNA (e.g. amino acids^[127], bile acids^[137], ferrocene^[130], tetrathiafulvalene^[138]). An overview is given in Figure 12.

Figure 12 Modified dATP derivatives for DNA polymerase incorporation. Examples for (a) terpyridine-, (b) phenylalanine-, (c) tetrathiafulvalene-, (d) cholic acid- and (e) azido- modifications.[127, 135-138]

(27)

Introduction 17

1.4 Diagnostic platforms

Since it is known that diseases are connected to genome variations; methods for sequencing or genotyping gained importance. At the same time platforms are needed that allow a high throughput of samples by using minimal amounts of DNA sample as well as high reproducibility. In the following two possible platforms are elucidated.

1.4.1 Microarrays

Microarrays are modern diagnostic tools that are able to analyze several properties in parallel by using only small amounts of sample material. For example, microarrays were already used to investigate protein-protein-, rezeptor-ligand or antibody-antigen-interactions.^[139] The surface material is variable and adapted to the application. Glass is eminently suited for microarrays: it is heat-stable, cheap and shows only little self-fluorescence. By now, glass-microarrays coated with many different functional groups (e.g. aldehyde-, epoxy-, and amine-coated surfaces) are commercially available. The target molecule can either be bound directly to the functional group or by the help of a bifunctional linker (for example: 1,4-phenylene diisothiocyanate).

In particular, DNA microarrays with their small format, high density and small probe volumes are an advantageous instrument for high throughput screening approaches.^[140]

The possibility of site-specific immobilization of primer probes enables the detection of many SNPs in parallel.

Employed methods for SNP detection are for example array based (e.g. SNP array systems from Affymetrix and Illumina). These arrays score with the ability to assess SNPs all over the genome and the copy number analysis on one chip simultaneously. For example each Genome-Wide Human SNP Array 6.0 (Affymetrix) carries 1.8 million genetic markers, including more than 900.000 SNPs and probes for the copy number analysis.^[141]

1.4.2 Streptavidin sepharose beads

Streptavidin sepharose material is often used to immobilize biotin (vitamin B7, vitamin H) or biotinylated substances through affinity interactions. Sepharose material is rigid, highly cross-linked beaded agarose with high chemical stability. On this material streptavidin is immobilized. Streptavidin is build up from four identical subunits. Each subunit can bind a biotin molecule with extraordinarily high affinity, so four biotin molecules can bind simultaneously to a single streptavidin molecule.^[142] The binding of streptavidin to biotin is one of the strongest known non-covalent interactions (Kd≈ 10^-14 mol/L^[143]). It is expected that several factors are responsible for the extraordinary affinity. First, Biotin fits perfectly into the binding pocket of streptavidin and forms several hydrogen bonds to residues in the binding site. Secondly, the dissociation of bound biotin is also hampered by

(28)

18 Introduction

a loop, which closes the binding pocket when biotin is bound.^[144] Hydrophobic interactions play also a crucial role for the affinity of binding.^[145]

Biotin can be easily attached to the 5’-end of an oligonucleotide during solid-phase automated synthesis. Several biotinylated phosphoramidites with different spacer types and lengths are commercially available. 5’-biotinylated DNA oligonucleotides are often used as primers in PCR to isolate one PCR product strand.^[146] Using a biotinylated primer in PCR the corresponding PCR product strand is biotinylated. The complementary strand remains unbiotinylated. By the addition of streptavidin-coated beads the biotinylated strand is captured on the bead and is separated from complementary unbiotinylated strand and impurities. This is a useful technique for purification and separation of one PCR product strand.

(29)

19 2 Aim of this work

The number of DNA-based technologies is rapidly growing. Many of these technologies depend on the incorporation of modified nucleotides into DNA catalyzed by DNA polymerases. For the efficiency of such applications, it is extremely important that the modified nucleotides are well accepted by the DNA polymerase. Hitherto, the mechanisms of acceptance and incorporation of modified substrates are still unclear and often not predictable. Learning more about the incorporation of modified nucleotides catalyzed by DNA polymerases will promote the rational design of modified nucleotides and this will lead to more powerful applications. Furthermore modified nucleotides with altered properties are needed to increase the scope of applications. Many nucleotide modifications are known nowadays and some are already successfully used in DNA applications like sequencing. The results of this doctoral thesis should contribute to several aspects of modified DNA:

• Increasing the understanding of the acceptance and incorporation mechanisms of modified nucleotides by DNA polymerases.

• The synthesis of modified purine nucleotides with new properties and their incorporation by DNA polymerases.

• The development of new methods of application based on the newly developed and synthesized modified nucleotides.

Therefore modified nucleotide surrogating the natural nucleotides should be synthesized first. Furthermore, their incorporation efficiency compared to their natural counterparts should be investigated. For comparability reasons all modified nucleotides should carry the same chain moiety. Mainly pentinyl-amine chains should be used, as many known nucleotide analogs carry an amine-linker moiety. In addition, the pentinyl-amine chain should be elongated with a extended alkyl-chain. To shed light on the incorporation process, crystal structures of KlenTaq DNA polymerase in complex with these pentinyl-amine-modified nucleotides should be prepared in collaboration with Dipl. Biol.

K. Bergen (University of Konstanz). The crystal structures should give detailed

(30)

20 Aim of this work

information about how modified nucleotides are processed and stabilized in the active site of the DNA polymerase.

To emphasize the possibility of further functionalization of these amine-modified nucleotides a label with new characteristics should be attached to the pentinyl-amine linker. Frequently used in biotechnological applications are modified nucleotides bearing fluorescent or affinity labels such as Cy3 or biotin. The newly developed label should have the potential to apply several detection techniques without changing the label.

Interestingly, DNA itself exhibit this advantage as the self-assembly and hybridization properties offer great potential. Several detection methodologies are conceivable by using these properties. Apart from all this positive things, one possible drawback can be the huge increase of size. A modified nucleotide bearing a label consisting of an oligodeoxynucleotide (ODN) is several times bigger than a natural nucleotide. Of course, this could impede the efficiency of incorporation by DNA polymerases. Therefore purine analogs carrying an ODN label should be synthesized. Additionally, in-vitro experiments should reveal the substrate scope and DNA polymerase incorporation efficiency of various DNA polymerases.

The diagnostic applicability of these ODN-modified nucleotides for genetic variation detection should be evaluated using common biotechnological platforms such as microarray technique or on-bead immobilization. The attached ODN-modification of the incorporated nucleotide should be used for the detection of the incorporation event. To detect even little amounts of incorporated unmodified nucleotides signal amplification is absolutely indispensable. Therefore, different strategies to amplify the obtained signal should be established. The microarray approach should allow detection on an ultra-high throughput level. In contrast, the on-bead approach should be as easy as possible, meaning that as less as possible equipment should be used. This should enhance the applicability and reduce the costs. Optimally this would lead to a naked eye detection system.

(31)

21 3 Results and Discussion

3.1 Modified nucleotides for the structural analysis of the incorporation mechanism by KlenTaq DNA polymerase

Crystallization experiments were designed, performed and evaluated by Dipl.-Biol. K.

Bergen (University of Konstanz) and B. Sc. S. Strütt (University of Konstanz). Parts of this chapter are already published in K. Bergen, A.-L. Steck,S. Strütt, A. Baccaro, W. Welte, K.

Diederichs, A. Marx J. Am. Chem. Soc. 2012, 134, 11840-11843^[147] and in the bachelor thesis of B. Sc. Stefan Strütt (2011, University of Konstanz). The syntheses of dT*TP, dT**TP and dC*TP were performed by Dr. A. Baccaro (University of Konstanz).

3.1.1 Introduction

DNA polymerases are the key enzymes in DNA replication and they also play a significant role in DNA damage repair.^{[71, 148]} Overall DNA polymerases are crucial for the stability and maintenance of the genetic information. But not only in all living cells, DNA polymerases are of paramount importance, due to their widespread applicability in PCR and sequencing DNA polymerases became an indispensable tool in biotechnological applications. During the last years, incorporation of nucleobase-modified 2’-deoxynucleoside-5’-O- triphosphates (dNTPs) became the essential step in many biotechnological applications.[110, 113, 149] For example, enzyme-mediated incorporation of fluorescently- labeled nucleotides is of significant importance in DNA sequencing approaches.[112, 113, 149, 150] Although the incorporation of modified nucleotides is widely employed, the mechanisms of acceptance and incorporation of modified substrates are still unclear and often not predictable.^[114] Getting inside into these mechanisms will open the door for rationally designing of modified nucleotides which should offer great opportunities for future applications. For an extensive understanding of DNA polymerase structure and function, crystallography is the method of choice. Several crystal structures of DNA polymerases in complex with natural or modified nucleotides have been published so far,

(32)

22 Results and Discussion

but a comprehensive study using modified surrogates for all natural nucleotides was still lacking. For better comparability, the surrogates should carry the same modification.

(33)

Results and Discussion 23

3.1.2 Results

In order to crystallize the N-terminally truncated form of the DNA polymerase I from Thermus aquaticus (KlenTaq) DNA polymerase with relevant modified substrates, amine-modified 2’-deoxynucleoside-5’-O-triphosphates³ were chosen (Figure 13). The introduced amine moiety is well-suited for further functionalization via amide bond formation and therefore these modified nucleotides are practicable precursor molecules for the introduction of other functional groups as shown by us^[151] (see also chapter 3.2).

The amine-modifications were introduced at the C5 position of pyrimidines and at the C7 position of 7-deazapurines (Figure 13). Modifications at these positions are favored, as the modification is directed into the major groove of the B-DNA helix and the Watson-Crick base pairing is not effected.[103, 124-126] In addition, nucleoside triphosphates carrying modifications at these positions have been accepted by DNA polymerases in multiple studies.[103, 125, 152-155]

Figure 13 Synthesis and structures of amine-modified nucleotides. A) Synthesis of the amine-modified deoxyadenosine analogs starting from 7-deaza-2’-deoxy-7-iodoadenosine. a) PdCl2(PPh3)2, CuI, Et3N, 2,2,2-trifluoro-N-(pent-4-ynyl)acetamide, DMF, RT, 75%; b) proton sponge, POCl3, TMP, 0°C, followed by nBu3N, Bis(tri-n-butylammonium)pyrophosphate, DMF, TEAB buffer and 33% NH3, 4%. B) Structure of dG*TP, dT*TP and dC*TP.

3 For better understanding, modified 2’-deoxyuridine derivatives are named modified thymidines dT*TP in this work.

(34)

24 Results and Discussion

A practicable and frequently used way to chemically introduce modifications at these positions is the Pd-catalyzed Sonogashira cross coupling of alkynes with the iodinated nucleosides followed by triphosphate formation.^[156-158] Based on this method the employed modified nucleotide analogs have been synthesized (Figure 13A).^[151] The iodinated nucleosides 7-deazaiodoadenosine and 7-deazaiodoguanosine were synthesized according to known procedures.[159, 160] Under Sonogashira conditions, the iodinated nucleosides were coupled with 2,2,2-trifluoro-N-(pent-4-ynyl)acetamide^[161]. After column chromatography and RP-MPLC the resulting nucleosides were isolated in appropriate yields. The phosphorylation of the nucleosides was performed according to a procedure described by T. Kovács and L. Ötvös^[157] – a modified version of the Yoshikawa method^[162]. In brief: In a one-pot reaction, the highly reactive 5’-phosphorodichloridate nucleoside is formed by adding electrophilic phosphorous oxychloride to the nucleoside dissolved in trimethylphosphate, followed by the addition of tri-n-butylammonia pyrophosphate in DMF and afterwards hydrolysation. During the reaction, hydrogen chloride is formed and is able to attack the unsaturated side chains of the modified nucleosides. To avoid this, proton sponge (1,8-bis(dimethylamino)naphthalene) is added to the reaction mixture to quench the formed protons. This reaction has a major advantage compared to a frequently used phosphorylation method developed by J. Ludwig and F. Eckstein.^[163] The phosphorylation is regioselective, accordingly the 2’-hydroxylgroup of the deoxyribose is not affected and so a 2’-protecting group is not necessary. Thus, reaction steps for installation and removal of the 2’-protecting group can be avoided. After the removal of the TFA protecting group by aqueous ammonia, the desired amine-modified nucleoside triphosphates were obtained.

To demonstrate that these amine-modified nucleoside triphosphates are further functionaliziable, dA*TP was reacted with an active ester to get dA**TP, which carries an extended chain modification. The reaction is depicted in Figure 14. The amine-modified dA*TP is dissolved in DMSO and an excess of succinimidyl 10-hydroxydecanoate^[164] was added. After 16 h the reaction mixture was lyophilized and the product was purified by preparative HPLC to give the extended chain-modified dA**TP. dT**TP carrying the extended chain modification was synthesized in the same way.

Figure 14 Structure and synthesis of dA**TP and dT**TP. A) Synthesis of dA**TP; a) succinimidyl 10-hydroxydecanoate, DMSO, RT, quantitative. B) Structure of dT**TP.

(35)

Results and Discussion 25

3.1.2.1 Incorporation of amine-modified dN*TPs and dN**TPs

In order to validate whether KlenTaq DNA polymerase is capable to incorporate these modified substrates dN*TP and dN**TP, single-nucleotide incorporation experiments were performed. A 24-nt primer with a ³²P label at the 5’ end and four different 36-nt templates, that code for the extension of the primer by a single complementary nucleotide, were used. It was found, that KlenTaq DNA polymerase is able to incorporate all tested modified nucleoside triphosphates comparable to the natural counterpart (see Figure 15B lane 9 and Figure 16 lane 9).

Figure 15 Competition experiments of natural nucleotides versus amine-modified nucleotides using KlenTaq DNA polymerase. A) Partial DNA sequences of primer 1 and template 1A employing dA*TP and dATP. B) PAGE analysis of an exemplary competition experiments employing KlenTaq DNA polymerase. The ratio of dA*TP/dATP was varied. Lane 0: 5’-³²P-labeled primer; lane 1: ratio:

0/1; lane 2: ratio: 1/10; lane 3: ratio: 1/4; lane 4: ratio: 1/2; lane 5: ratio: 1/1; lane 6: ratio: 2/1;

lane 7: ratio: 4/1; lane 8: ratio: 10/1; lane 9: ratio: 1/0. C) Evaluation of the incorporation efficiency using mixtures with varied composition of dA*TP (■, dashed line) and dATP (•, solid line) and KlenTaq DNA polymerase. The conversion in % is plotted versus the ratio. The dotted line marks the approximate ratio where both nucleotides are equally incorporated. D) Overview of the efficiencies of the presented modified nucleotides in competition with their natural counterparts.

26 Results and Discussion

Figure 16 Competition experiments of natural nucleotides versus amine-modified nucleotides using KlenTaq DNA polymerase. A) Partial DNA sequences of primer 1 and template 1T employing dT*TP/dTTP and PAGE analysis of the competition experiments. The ratio of dT*TP/dTTP was varied. Lane 0: 5’-³²P-labeled primer; lane 1: ratio: 0/1; lane 2: ratio: 1/10; lane 3: ratio: 1/4; lane 4:

ratio: 1/2; lane 5: ratio: 1/1; lane 6: ratio: 2/1; lane 7: ratio: 4/1; lane 8: ratio: 10/1; lane 9: ratio:

1/0. B) Partial DNA sequences of primer 1 and template 1C employing dC*TP/dCTP and PAGE analysis of the competition experiments. The ratio of dC*TP/dCTP was varied. Lane 0: 5’-³²P- labeled primer; lane 1: ratio: 0/1; lane 2: ratio: 1/10; lane 3: ratio: 1/4; lane 4: ratio: 1/2; lane 5:

ratio: 1/1; lane 6: ratio: 2/1; lane 7: ratio: 4/1; lane 8: ratio: 10/1; lane 9: ratio: 1/0. C) Partial DNA sequences of primer 1 and template 1G employing dG*TP/dGTP and PAGE analysis of the

competition experiments. The ratio of dG*TP/dGTP was varied. Lane 0: 5’-³²P-labeled primer; lane 1: ratio: 0/1; lane 2: ratio: 1/10; lane 3: ratio: 1/4; lane 4: ratio: 1/2; lane 5: ratio: 1/1; lane 6: ratio:

2/1; lane 7: ratio: 4/1; lane 8: ratio: 10/1; lane 9: ratio: 1/0. D) Partial DNA sequences of primer 1 and template 1T employing dT**TP/dTTP and PAGE analysis of the competition experiments. The ratio of dT**TP/dTTP was varied. Lane 0: 5’-³²P-labeled primer; lane 1: ratio: 0/1; lane 2: ratio:

1/10; lane 3: ratio: 1/4; lane 4: ratio: 1/2; lane 5: ratio: 1/1; lane 6: ratio: 2/1; lane 7: ratio: 4/1;

lane 8: ratio: 10/1; lane 9: ratio: 1/0. E) Partial DNA sequences of primer 1 and template 1A employing dA**TP/dATP and PAGE analysis of the competition experiments. The ratio of

dA**TP/dATP was varied. Lane 0: 5’-³²P-labeled primer; lane 1: ratio: 0/1; lane 2: ratio: 1/10; lane 3: ratio: 1/4; lane 4: ratio: 1/2; lane 5: ratio: 1/1; lane 6: ratio: 2/1; lane 7: ratio: 4/1; lane 8: ratio:

10/1; lane 9: ratio: 1/0.

(37)

Results and Discussion 27

Figure 17 Competition experiments of dN*TP/dN**TP (■, dashed line) versus dNTP (•, solid line) using KlenTaq DNA polymerase. The conversion in % was plotted versus the concentration. The dotted line marks the approximate ratio where both nucleotides are equally incorporated.

A) Partial DNA sequences of primer 1 and template 1T for the incorporation of dT*TP and dTTP and evaluation of the reactions. B) Partial DNA sequences of primer 1 and template 1C for the incorporation of dC*TP and dCTP and evaluation of the reactions. C) Partial DNA sequences of primer 1 and template 1G for the incorporation of dG*TP and dGTP and evaluation of the reactions.

D) Partial DNA sequences of primer 1 and template 1A for the incorporation of dA**TP and dATP and evaluation of the reactions. E) Partial DNA sequences of primer 1 and template 1T for the incorporation of dT**TP and dTTP and evaluation of the reactions.

KlenTaq DNA polymerase incorporates the purine analogs with approximately the same efficiency as their natural counterparts, whereas the pyrimidine analogs were incorporated with about 12 to 34-fold lower efficiency than their natural counterparts (Figure 15D). Compared to dA*TP and dT*TP, the nucleotides with extended chain modification dA**TP and dT**TP were incorporated with higher efficiency.